從中醫內科病例報告看聊天機器人（ChatGPT）對中醫臨床教育之影響

Examining the Impact of ChatGPT on Traditional Chinese Medicine Clinical Education through Case Reports from Internal Medicine

10.7102/JTCIM.202312_21(2).0002

陳瑩陵(Ying-Ling Chen)；侯俊成(Chun-Cheng Hou)；許雅雰(Ya-Fen Hsu)；張順昌(Shun-Chang Chang)；李聰界(Tsung-Chieh Li)

病例報告 ； 聊天機器人 ； 醫學教育 ； case reports ； ChatGPT ； medical education

中醫內科醫學雜誌

21卷2期（2023 / 12 / 01）

11 - 17

繁體中文;英文

ChatGPT是一項新的科技工具，已經廣泛被人們所使用，其影響仍難以預估。中醫學生必然會將其運用在臨床學習上。本研究擬透過學生所撰寫之病例報告與ChatGPT所製作之病例報告比較其差異，以對未來中醫臨床教育預作準備。從2022年中醫實習醫學生所繳交之內科病例報告隨機抽選14份，去連結後擷取病例之四診資料，輸入ChatGPT產生相同格式之病例報告。再由3位具有中醫內科專科醫師指導資格之中醫臨床教師，對每份報告進行量化評估，再將所得之評估數據進行統計分析，檢定學生所完成之病例與ChatGPT完成之病例的差異。結果發現除了基本病史紀錄外，在病因病機分析、診斷、處方及辨證正確性，學生皆優於聊天機器人（p<0.05）。因此聊天機器人在現階段還不會對中醫病例報告學習產生巨大影響。

ChatGPT is a new technological tool that has been widely used, with impacts that are still hard to predict. Students of traditional Chinese medicine will inevitably apply it to their clinical learning. This study aims to prepare for the future of traditional Chinese medicine clinical education by comparing the differences between case reports written by students and those produced by ChatGPT. Fourteen internal medicine case reports submitted by interns in traditional Chinese medicine in 2022 were randomly selected, the diagnostic data of the cases were extracted and entered into ChatGPT to generate case reports in the same format. Three clinical teachers with qualifications to guide specialists in traditional Chinese medicine internal medicine then conducted quantitative evaluations of each report. The evaluation data were statistically analyzed to determine the differences between the cases completed by the students and those completed by ChatGPT. The results showed that, except for basic medical history records, students outperformed the chatbot in terms of disease cause and mechanism analysis, diagnosis, prescription, and correctness of syndrome differentiation (p<0.05). Therefore, at this stage, chatGPT will not have a significant impact on the learning of traditional Chinese medicine case reports.

1. Doshi, R.,Amin, K.,Khosla, P.,Bajaj, S.,Chheang, S.,Forman, H. P.(2023).,未出版
2. Khan, R. A.,Jawaid, M.,Khan, A. R.,Sajjad, M.(2023).ChatGPT-Reshaping medical education and clinical management.Pakistan Journal of Medical Sciences,39(2),605.
3. Lin, J. W.(2023).Acupuncture Points in Traditional Chinese Medicine: A Medical Treatment Accessing to Particularly Good Pressing Effect Using ChatGPT.European Journal of Medical and Health Sciences,5(2),58-60.
4. Rosol, M.,Gasior, J. S.,Laba, J.,Korzeniewski, K.,Mlynczak, M.(2023).,未出版
5. Takagi, S.,Watari, T.,Erabi, A.,Sakaguchi, K.(2023).Performance of GPT-3.5 and GPT-4 on the Japanese Medical Licensing Examination: Comparison Study.JMIR Medical Education,9(1),e48002.
6. 王怡心,劉育明(2023)。數位審計新方向：監督，洞察，前瞻之情境分析。會計研究月刊，450，106-113。
7. 林宏孟(2023)。淡江大學，淡江大學碩士論文。